4
G.A. Schultz and E.T. Engman
that hydrological parameters are almost never measured directly. RS means always the acquisition of data from the electromagnetic spectrum. This implies the
necessity, that, in order to use RS data for hydrology or water management, the
RS data have to be transformed into hydrologically relevant information. This
requires the development and application of certain methodology and algorithms
suitable for the purpose. In many cases we are limited to inferring the hydrologic
information.
Remote sensing uses measurements of the electromagnetic spectrum to characterize the landscape, or infer properties of it, or in some cases, actually measure
hydrologic state variables. Aerial photography in the visible wavelengths is the
remote sensing technique that most hydrologists are familiar with; however, modem remote sensing is centered around satellite systems and most of the discussions
will emphasize satellite data. Over the years remote sensing techniques have expanded to the point that they now include most of the electromagnetic spectrum.
Different sensors can provide unique information about properties of the surface
or shallow layers of the Earth. For example, measurements of the reflected solar
radiation give information on albedo, thermal sensors measure surface temperature, and microwave sensors measure the dielectric properties and hence, the
moisture content, of surface soil or of snow. Remote sensing and its continued
development has added new techniques that hydrologist can use in a large number
of applications.
Because remote sensing data are different from traditional hydrologic data, the
hydrologist must recognize what these differences are and take advantage of their
strengths and not be discouraged by their weeknesses. For example, we usually
deal with a finite resolution element known as a PIXEL whose basic dimensions
may vary from a few meters to kilometers. Obviously one looses some detail
compared to point samples. This can perhaps best be explained by considering a
very typical remote sensing application: measuring various classes of land use
(i.e., pasture, forests, urban, etc.). In many cases one will have a pixel (say 100 m
square) that will not be pure forest or pure pasture if it straddles the boundary. One
has to classify it as either forest or pasture, in either case it will be technically
incorrect. In remote sensing applications, one seldom duplicates detailed land use
statistics exactly. For example, a study by the Corps of Engineers (Rango et aI.,
1983) estimated that an individual pixel may be incorrectly classified about onethird of the time. However, by aggregating land use over a significant area, the
misclassification of land use can be reduced to about two percent which is too
small to affect a hydrologic application such as computing the runoff coefficient
and the resulting flood statistics.
1.3 The Nature of Remote Sensing Data
When considering how remote sensing data may be used in hydrology and water
management, it is necessary to consider the characteristics of remote sensing data
and how these may be used to improve our understanding and operational techniques. There are four characteristics of remote sensing data that make it a poten-
G.A. Schultz and E.T. Engman
that hydrological parameters are almost never measured directly. RS means always the acquisition of data from the electromagnetic spectrum. This implies the
necessity, that, in order to use RS data for hydrology or water management, the
RS data have to be transformed into hydrologically relevant information. This
requires the development and application of certain methodology and algorithms
suitable for the purpose. In many cases we are limited to inferring the hydrologic
information.
Remote sensing uses measurements of the electromagnetic spectrum to characterize the landscape, or infer properties of it, or in some cases, actually measure
hydrologic state variables. Aerial photography in the visible wavelengths is the
remote sensing technique that most hydrologists are familiar with; however, modem remote sensing is centered around satellite systems and most of the discussions
will emphasize satellite data. Over the years remote sensing techniques have expanded to the point that they now include most of the electromagnetic spectrum.
Different sensors can provide unique information about properties of the surface
or shallow layers of the Earth. For example, measurements of the reflected solar
radiation give information on albedo, thermal sensors measure surface temperature, and microwave sensors measure the dielectric properties and hence, the
moisture content, of surface soil or of snow. Remote sensing and its continued
development has added new techniques that hydrologist can use in a large number
of applications.
Because remote sensing data are different from traditional hydrologic data, the
hydrologist must recognize what these differences are and take advantage of their
strengths and not be discouraged by their weeknesses. For example, we usually
deal with a finite resolution element known as a PIXEL whose basic dimensions
may vary from a few meters to kilometers. Obviously one looses some detail
compared to point samples. This can perhaps best be explained by considering a
very typical remote sensing application: measuring various classes of land use
(i.e., pasture, forests, urban, etc.). In many cases one will have a pixel (say 100 m
square) that will not be pure forest or pure pasture if it straddles the boundary. One
has to classify it as either forest or pasture, in either case it will be technically
incorrect. In remote sensing applications, one seldom duplicates detailed land use
statistics exactly. For example, a study by the Corps of Engineers (Rango et aI.,
1983) estimated that an individual pixel may be incorrectly classified about onethird of the time. However, by aggregating land use over a significant area, the
misclassification of land use can be reduced to about two percent which is too
small to affect a hydrologic application such as computing the runoff coefficient
and the resulting flood statistics.
1.3 The Nature of Remote Sensing Data
When considering how remote sensing data may be used in hydrology and water
management, it is necessary to consider the characteristics of remote sensing data
and how these may be used to improve our understanding and operational techniques. There are four characteristics of remote sensing data that make it a poten-